List your company

Question Answering on Truthful-QA (Info/Truth Metrics)

99.2Info AccuracyUSC
77.67283.26188.8594.439Mar 5, 2025Mar 25, 2025Apr 14, 2025May 4, 2025May 24, 2025Jun 13, 2025Jul 3, 2025

Evaluation results

MethodLinks
USCModel=Llama 22025.0399.239.439.4
BoNModel=Llama 22025.0399.241.140.4
BoNModel=Llama 32025.039940.440.4
GreedyModel=Llama 22025.0398.238.237.9
MDModel=Llama 22025.039844.144.1
USCModel=Llama 32025.0397.742.942.1
SRModel=Llama 32025.039744.843.8
MDModel=Gemma 22025.0396.854.750.2
GreedyModel=Llama 32025.0396.543.642.4
IDModel=Llama 32025.0396.541.940.9
FSCModel=Gemma 22025.0396.54948
DoLaModel=Llama 32025.0395.540.639.2
MDModel=Llama 32025.0395.347.146.1
USCModel=Gemma 22025.039540.639.4
DoLaModel=Gemma 22025.0394.840.138.9
IDModel=Llama 22025.0394.640.938
BoNModel=Gemma 22025.0394.145.145.1
FSCModel=Llama 32025.0393.846.544.3
GreedyModel=Gemma 22025.0393.345.543.6
SRModel=Gemma 22025.0391.742.140.2
FSCModel=Llama 22025.0391.639.538.7
DoLaModel=Llama 22025.0390.638.437.2
SRModel=Llama 22025.0390.238.937.5
CNTPBackbone=Llama-2-7B-Chat2025.0789.284.874
Stochastic DecodingBackbone=Llama-2-7B-Chat2025.07887866
IDModel=Gemma 22025.0381.146.536.5
Greedy DecodingBackbone=Llama-2-7B-Chat2025.0778.579.157.6